See bottom for continuation of discussion.
On 11/19/2015 2:19 PM, Ken Springer wrote:
On 11/19/15 11:04 AM, Paul D. Mirowsky wrote:
A bit of a rant. My apologies and respect to all who have made Draw
possible in advance.
On 11/19/2015 11:25 AM, Regina Henschel wrote:
Hi Dave,
dave boland schrieb:
Thanks all. If I make a suggestion to LO - add some more CAD
abilities
to Draw such as higher precision.
The current internal precision is 1/100 mm. Armin Le Grand had started
to change Draw to number format double to allow higher precision, see
https://wiki.openoffice.org/wiki/Aw080_documentation. Unfortunately
there is currently no Company which will pay for finishing that work.
Thank you for pointing this out.
You can get a very high precision in the UI, if you view the drawing
with 3000% zoom and move/scale the objects using the array keys
together with the Alt-key.
If you insist on the inch system, and need fractions of inches, you
should change the unit from inch to point. It is 72 point = 1 inch, so
0.125 inch can be entered as 9 point, and with such number there is no
problem in the UI. But you should notice that 1 inch = 2.54 cm and
therefore 0.125 inch = 0.3175 cm = 317.5 * 1/100mm. It cannot be
represented exactly with the internal 1/100mm unit.
If I insist on the inch system, I should use inches. Not meant to be
harsh, but really?
Can't debate your math, but shouldn't data be entered as it is measured?
For discussion's sake only, not to start an argument, if you are going
to measure in inches as your selection, then to expand on that, would
you want your font to be measured in inches instead of points?
<snip>
All measurement is application specific, using the speed of light as an
interpretation of inches or fonts, although accurate, is not reasonable
for CPU cycles.
It is very much like the 'epoch' date problem. The underlying math is
singular, but results are interpreted as seconds, days, months that we
use are not.
The "Today" function is derived from the 'Epoch' calculations so it is
understood by anyone using it. It is not loose, but reasonable within
expectation.
To think in other manner may not be folly, but it will still result in
inconsistency.
Today, we still have the mathematical 'rounding' problem that results in
different accounting results.
(Hmmm, where did the thousand dollars go in that billion dollar transaction)
Compare Calcs' underlying method for time and date and Draws' underlying
method for measurements.
As an example of one method, when a page is created, the first point is
created which becomes a reference for any other position using A2 + B2 =
C2 within X Y Z co-ordinates and it's "measurement" is mathematically
derived into inches, pica's, millimeters, etc.. While Draw must use
basic math correctly underneath, the interpreted results in measurement
do not follow functionally.
This issue recognized, I still love using it and LibreOffice.
Ref: Calc for Date & Time Functions for Calc at
https://help.libreoffice.org/Calc/Date_and_Time_Functions
Ref: Epoch date: January 1, 1970 --- Unix Epoch aka POSIX time, used by
Unix and Unix-like systems (Linux, Mac OS X), and programming languages:
most C/C++ implementations, Java, JavaScript, Perl, PHP, Python, Ruby,
Tcl, ActionScript. Also used by Precision Time Protocol.
Ref: Pythagorean theorem at
https://en.wikipedia.org/wiki/Pythagorean_theorem
--
To unsubscribe e-mail to: [email protected]
Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://listarchives.libreoffice.org/global/users/
All messages sent to this list will be publicly archived and cannot be deleted
